Method and apparatus for transmitting LTE waveforms in a shared spectrum with carrier sensing - Patents.com

Abstract

To provide a procedure for handling a hidden terminal in CSMA / CA on a shared frequency spectrum.SOLUTION: In a long term evolution (LTE) communication system on a shared frequency spectrum, user equipment (UE) is initialized (400) on an LTE frequency band. A base station (eNB) monitors the shared frequency spectrum (402) to determine if the share frequency spectrum is BUSY or not (404). The eNB transmits to the UE on the shared frequency spectrum if the shared frequency spectrum is not BUSY (406). The eNB waits for a first time (408) if the shared frequency spectrum is BUSY and instructs the UE to release the shared frequency spectrum after the first time (410).SELECTED DRAWING: Figure 4A

Description

[Technical field]

[0001] In most countries, access to the radio frequency spectrum is heavily regulated by government agencies, such as the Federal Communications Commission (FCC) in the United States and the European Commission in the European Union. Like any other natural resource, frequencies in the radio spectrum must be shared among its users. Thus, portions of the radio spectrum (e.g., bands) are either licensed to individual users (e.g., cell phone operators) or shared among multiple users, such as WiFi or Bluetooth, which operate in unlicensed bands. In some hybrid models, licensed spectrum is given to a primary user with the highest priority (e.g., naval radar applications). Secondary users can also use the licensed band during periods of inactivity when the primary user is not transmitting waveforms in that band. These secondary users may have different priorities. For example, a given frequency band licensed to a primary user may be used by a public safety organization for mission-critical communications. In this case, commercial users may be allowed to use such bands, but only if neither the primary nor the higher priority secondary users (e.g., public safety users) occupy the band. Such policy-based spectrum use is sometimes referred to as authorized shared access (ASA). From this point of view, there is no need to distinguish between unlicensed and licensed shared access, since these same techniques can be used to ensure fairness and policy compliance whenever a bandwidth is used by multiple users.

[0002] In the above example of licensed shared access, spectrum sharing can be facilitated by dynamic methods, sometimes referred to as listen-before-talk (LBT) methods, and by semi-static methods, such as geolocation databases (GLDBs). For example, such databases can map frequency usage in a band to geographical regions or time periods of a day. These databases cannot change dynamically, as a period is required to update and propagate these databases to all relevant users. LBT methods, as the name suggests, are more dynamic and do not rely on semi-statically configured databases. Instead, secondary users must ensure that primary users or other users of equal priority are not disturbed by secondary users' transmissions. Two well-known examples are radar avoidance and carrier sense carrier sense multiple access with collision avoidance (CSMA / CA) in IEEE 802.11 wireless local area networks (WLANs). The former applies when secondary users must give priority to primary users. This is sometimes called Dynamic Frequency Selection (DFS) because the secondary user must cease transmission upon detection of a military, weather, or automotive radar waveform. Thus, the secondary user frees up a given band or channel (a channel is a further division of a band) upon detection of a primary user and attempts to transmit on a different band or channel, hence the name Dynamic Frequency Selection. Similarly, in CSMA / CA, if the transmitter detects an ongoing transmission of the same priority, it chooses not to transmit in order to try again at a later time, hence the name Carrier Sense Multiple Access with Collision Avoidance. Thus, the main differences between DFS and CSMA / CA are the time scale on which sensing occurs and the action taken by the transmitter when an ongoing transmission is detected. For example, a DFS transmitter must always switch channels / bands to free up the current channel or band for the primary user, whereas a CSMA / CA transmitter may or may not switch channels.This is because in CSMA / CA, radio resources are shared among users of the same priority, which is considered a multiple access scheme. However, with DFS, the primary user has a higher priority. As a result, to guarantee the contention latency of the CSMA / CA scheme, carrier sense (CS) and collision avoidance (CA) occur on the order of tens of microseconds (μs), whereas DFS can take several seconds.

[0003] The operation of the CS / CA multiple access scheme is in stark contrast to other common multiple access techniques such as time division multiple access (TDMA), frequency division multiple access (FDMA), code division multiple access (CDMA), or orthogonal frequency division multiple access (OFDMA). This is due to the nature of opportunistic random access to share the medium. TDMA and FDMA in GSM, CDMA in UMTS, and orthogonal frequency division multiple access (OFDMA) in the Third Generation Partnership Project (3GPP) Long Term Evolution (LTE) attempt to orthogonalize available resources for sharing among multiple users. However, the orthogonalization operation requires precise coordination by predefined rules or dynamic schedulers, which assign resources to specific users for a given period in a given portion of the radio frequency spectrum, so that collisions are essentially avoided. This orthogonalization operation makes it particularly difficult to operate these resources in radio resources shared by the CA / CA multiple access scheme. This is because, when competing for available radio resources, users following these types of protocols and procedures yield to users following predefined schedules or radio resource allocations according to these protocols and procedures.

[0004] In LTE, a base station is known as an evolved NodeB (eNodeB / eNB) and has full control over the radio resource management (RRM) of the cells under LTE control. An E-UTRAN (Evolved Universal Terrestrial Radio Access Network) typically includes many eNodeBs, each with its own RRM functionality. A subset of these eNodeBs may coordinate their RRM via an X2 Application Protocol (X2AP) defined on an X2 interface connecting two eNodeBs. Similarly, each eNodeB is connected to one or more of the Mobility Management Entities (MMEs) in the Core Network (CN) via an S1 interface, where an S1 Application Protocol (S1AP) is defined. The S1AP may also be used for RRM coordination. The RRM interface is an essential part of cellular communication, as it enables important functions such as interference coordination, mobility, and even Self-Organizing Network (SON).

[0005] 1 is an example of a prior art wireless long range communication network. The illustrated long range communication network includes a primary eNodeB 110 operating in a primary cell (PCell) 100 and eNodeBs 112, 114, 116, and 118 operating in secondary cells (SCell1-SCell4) 102, 104, 106, and 108. A handset or other user equipment (UE) 120 is shown communicating with the eNodeB 110 of the PCell 100. The UE 120 may also communicate with one or more eNodeBs of the secondary cells. In this example, the SCell is a logical concept, and thus the eNodeB 110 may operate multiple SCells 102-108.

[0006] The eNodeB 110 also controls the radio resources in its cell 100 via a Radio Resource Control (RRC) protocol and also the multiple access of users connected to the cell via a Medium Access Control (MAC) protocol. For example, the RRC protocol configures the carriers on which a user equipment (UE) may transmit and receive data, and up to five so-called component carriers (CCs) may be configured per UE in LTE-Advanced (LTE-A). Similarly, the MAC protocol together with the RRC protocol controls how and when the UE may use the available radio resources to transmit and receive data on the configured carriers. In LTE Release 10, a feature called carrier aggregation is introduced. In carrier aggregation, a UE may be configured with one primary cell (PCell) and up to four secondary cells (SCells). The PCell may only be changed by handover, whereas the SCells are configured by RRC signaling. In particular, the UE is not expected to receive system information by decoding a Physical Broadcast Channel (PBCH) on a Secondary Component Carrier (SCC) or to receive System Information (SI) on a Downlink Shared Channel (DL-SCH) by monitoring a common search space of the SCell to receive a Physical Downlink Control Channel (PDCCH) whose CRC is scrambled by the SI-RNTI. Also, the UE may assume that the System Frame Number (SFN) on all SCCs aligns with the SFN of the Primary Component Carrier (PCC).

[0007] CA does not define Radio Link Monitoring (RLM) of the SCell. Thus, there is no specified means for the UE physical layer (PHY) to indicate Radio Link Failure (RLF) to upper layers of the UE via the MAC layer. This is because in E-UTRA (Evolved Universal Terrestrial Radio Access), one can always rely on the connectivity provided by the PCell, which provides robustness through RLM and other fallback procedures. Alternatively, the SCell acts as a supplementary serving cell that can be activated when additional capacity is required for data communication with the UE. For this purpose, the MAC layer can activate a configured SCell via the MAC Control Element (CE). Activating the SCell can require 8 to 30 ms depending on the synchronization state of the UE with its SCC. RRC reconfiguration of the SCell can take longer than this, especially if the UE needs to perform intermediate frequency measurements. Therefore, the eNodeB may configure the UE to periodically measure the Reference Signal Received Power (RSRP) of a cell on a carrier and report the measurements periodically or triggered by a configurable offset and threshold. In 3GPP Long Term Evolution, this is achieved via RRC signaling of measurement objects and configurations. If measurements are readily available at the eNodeB, the latency of RRC reconfiguration of SCell or PCell is dramatically reduced from a few seconds to tens or hundreds of milliseconds. The eNodeB activates only cells already configured as SCells, but the eNodeB may configure the UE to measure RSRP in any cell. In comparison, the eNodeB may use the measurement reports of any cell to activate a cell, as in the case of activating a SCell, or to RRC reconfigure the UE to add / remove a SCell, or even to change the PCell.

[0008] After a PCell or SCellS is activated, the eNodeB MAC scheduler assigns downlink (DL) and uplink (UL) grants to the UE for downlink and uplink transmissions on the Physical Downlink Shared Channel (PDSCH) and Physical Uplink Shared Channel (PUSCH), respectively. In the downlink direction, a grant received in the Downlink Control Information (DCI) in subframe n schedules a corresponding PDSCH transmission in the same subframe. In comparison, in the uplink, this grant schedules a PUSCH transmission in subframe n+k, where k>0 is determined by a pre-specified rule.

[0009] The E-UTRAN (especially the eNodeB) has full control of all radio resources, at least for UEs in RRC_CONNECTED mode. Except for the Physical Random Access Channel (PRACH), the E-UTRAN controls all transmissions in both uplink and downlink directions, including resource allocation by time, frequency, or any other means such as CDMA, as well as timing and power control of transmissions.

[0010] The eNodeB has an RRM function, which controls all radio resources via RRC, but relies on the UE to discover cells and report related measurements. To this end, in LTE Releases 8-11, the eNodeB transmits in each radio frame a Primary Synchronization Signal (PSS), a Secondary Cell Synchronization Signal (SSS), and a Cell-Specific Reference Signal (CRS). While the PSS and SSS each occupy one OFDM symbol per half frame, the CRS is transmitted in each subframe of the radio frame, so that the UE can discover and measure cells within a 6 ms measurement window, even if the timing of a given cell is not known a priori. Also, to support inter-frequency measurements in Time Division Duplex (TDD) systems where the UL / DL configuration of the cell may be unknown to the UE, or to support measurement constraints introduced in LTE Release 10 for the purpose of enhanced Inter-cell Interface Coordination (eICIC), the UE needs to be able to discover cells in only one subframe, and possibly in the DwPTS field of a special subframe. To facilitate energy saving and interference reduction, LTE Release 12 introduces a "discovery burst (including PSS, SSS, and CRS transmissions)" and, if configured, a channel state information reference signal (CSI-RS) for transmission point (TP) identification in the context of a shared cell ID. For example, multiple TPs may share the same physical cell ID and can only be distinguished by their respective CSI-RS resource element (RE) configurations. The PSS, SSS, CRS, and CSI-RS (if configured) form a discovery reference signal (DRS) and are transmitted during DRS occasions. DRS occasions are similar to LTE Release 9 positioning reference signal (PRS) occasions, as they have a configured or specific length (e.g., number of subframes) and periodicity. Ideally, the length of a DRS occasion is no longer than the UE measurement window of 6 ms and can be as short as one subframe.A reasonable periodicity for DRS occasions is several hundred milliseconds, and DRS bursts may act as beacons in other wireless communication systems (such as CSMA / CA). Summary of the Invention

[0011] In a first embodiment, a method for operating a Long Term Evolution (LTE) communication system on a shared frequency spectrum is disclosed. A base station (eNB) initializes a user equipment (UE) on an LTE frequency band. The base station (eNB) monitors the shared frequency spectrum to determine whether the shared frequency spectrum is BUSY. If the shared frequency spectrum is not BUSY, the eNB transmits to the UE on the shared frequency spectrum. If the shared frequency spectrum is BUSY, the eNB waits for a first time. After the first time, the eNB instructs the UE to release the shared frequency spectrum.

[0012] In a second embodiment, the UE monitors the shared frequency spectrum to determine whether the shared frequency spectrum is BUSY. If the shared frequency spectrum is not BUSY, the UE transmits to the eNB on the shared frequency spectrum. If the shared frequency spectrum is BUSY, the UE waits for a first time. After the first time, the UE reports a BUSY state to the eNB. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a diagram of a prior art Long Term Evolution (LTE) communication system.

[0014] [Diagram 2] 1 is a flow chart illustrating operation of a Long Term Evolution (LTE) communication system in a Licensed Shared Access (ASA) frequency spectrum.

[0015] [Diagram 3] FIG. 2 illustrates communication between a user equipment (UE) and a base station (eNB) according to an example embodiment.

[0016] [Figure 4A] 4 is a flow chart illustrating downlink operation of a Long Term Evolution (LTE) communication system in a Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA) frequency spectrum in accordance with an example embodiment.

[0017] [Figure 4B] 1 is a flow chart illustrating uplink operation of a Long Term Evolution (LTE) communication system in a Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA) frequency spectrum in accordance with an example embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0018] Exemplary embodiments are directed to an apparatus and method for operating an Orthogonal Frequency Division Multiple Access (OFDMA) cellular communications system, such as 3GGP Long Term Evolution (LTE), on a radio frequency shared with a primary transceiver. The primary transceiver may be a naval, mobile radio, or other transceiver having a high priority. Although specific terminology is used herein, it is used in a generic and descriptive sense only and not for purposes of limitation. The following abbreviations are used throughout this specification: ASA: Authorized Shared Access eNB: evolved Node B or base station UE: User Equipment CQI: Channel Quality Indicator CRS: Cell-specific reference signal CSI: Channel State Information CSI-RS: Channel State Information Reference Signal CSMA / CA: Carrier Sense Multiple Access with Collision Avoidance DCI: Downlink control information DFS: Dynamic Frequency Selection DRS: Discovery Reference Signal DL: Downlink DwPTS: Downlink pilot time slot E-UTRAN: Evolved Universal Terrestrial Radio Access Network LBT: Listen Before Talk LTE: Long Term Evolution MAC: Medium Access Control Protocol MIMO: Multiple Input Multiple Output OFDMA: Orthogonal Frequency Division Multiple Access OOR: Out Of Range PBCH: Physical Broadcast Channel PCell: Primary cell PCFICH: Physical Control Format Indicator Channel PDCCH: Physical Downlink Control Channel PDSCH: Physical Downlink Shared Channel PHICH: Physical Hybrid ARQ Indicator Channel PMCH: Physical Multicast Channel PSS: Primary Synchronization Signal PUCCH: Physical uplink control channel PUSCH: Physical uplink shared channel RI: Rank Indicator RRC: Radio Resource Control RRM: Radio Resource Management RSRP: Reference Signal Received Power SCell: Secondary cell SRS: Sounding Reference Signal SSS: Secondary Synchronization Signal TDD: Time division duplex TRS: Tracking Reference Signal UL: Uplink

[0019] <Dynamic Frequency Selection (DFS)> It is not easy to use the 3GPP Long Term Evolution (LTE) communication standard in a shared access spectrum, because the radio resource management function resides in the eNodeBs in the network and the radio resources are controlled only by these eNodeBs. The Dynamic Frequency Selection (DFS) method typically provides enough time (e.g., several seconds) to change the frequency band or carrier after the primary user is detected. Therefore, handover-based RRC signaling and activation or deactivation of SCell under MAC control is sufficient to release the band to the primary user. The 3GPP LTE communication standard currently lacks protocols, procedures, and measurements that make the UE take some action when a primary user is detected on the carrier on which the UE is configured to transmit data. Also, mobility control in LTE is fully controlled by the eNodeB, while other wireless cellular communication standards make the UE initiate a handover. Here, mobility balances the load, in which the eNodeB may add or remove SCells or change the PCell for stationary UEs. In both the ASA-based scheme with a primary user and the CSMA / CA-based scheme without a primary user, so-called "hidden stations" may exist. Hidden stations are transmitters, such as primary users, whose transmissions can only be detected at the receiving end of a communication link sharing the wireless medium. For example, in LTE, only the UE can detect the waveform transmitted from a "hidden station," but the eNodeB is completely unaware of its existence.

[0020] FIG. 2 is a flow chart showing the operation of the first embodiment. In step 200, a UE is initialized to operate in an LTE band with a PCell. The ASA band is configured and operated by the eNodeB as a regular LTE band, and the UE operates in the ASA band (202). The UE is prohibited from camping on a cell operating in the ASA band by existing means, such as prohibition by broadcasting system information. As a result, all UEs connected in the ASA band are in RRC_CONNECTED mode, and are thereby under the full control of the eNodeB. The eNodeB configures all UEs connected in the ASA band to perform RRM measurements according to existing LTE specifications (e.g., Releases 8-12) (204). DFS is supported by each UE by a non-standard (proprietary) implementation. When the UE detects a hidden terminal (from the UE's perspective, all primary users are hidden terminals) (206), the UE's physical layer (PHY) indicates the detected hidden terminal to the upper layers of its protocol stack and triggers an RRM measurement report according to existing LTE Release 8 / 9 / 10 / 11 / 12 procedures. Depending on the specification (e.g., "DFS event"), the RRM measurement report triggered by the non-standard (proprietary) DFS function in the UE may be tied to a specific value of the RRC information element (IE) RSRP range. For example, a DFS event may be indicated by the minimum value in the RRC IE RSRP range and may act as an out-of-range (OOR) indication. The UE may report the DFS event (e.g., an RSRP measurement report with an OOR indicator signaling the DFS event) to the eNodeB using existing RRM measurement reporting procedures (208). The eNodeB's RRM function may reinterpret the RSRP measurement report as a DFS / OOR event according to the standardized linkage and subsequently reconfigure the UE via existing RRC signaling 212 to release the ASA band to the primary user 210. Such RRC signaling includes handover in the case of a PCell or SCell reconfiguration in the case of an SCell.Alternatively, if the RRM function in the eNodeB determines that the ASA band should be temporarily released to the primary user, the RRM function may simply let the sCell_Deactivation_Timer in the UE expire, or may send a deactivation command in the MAC Control element to deactivate the SCell configured in the ASA band. Although the 3GPP LTE specifications may introduce performance requirements that are useful to test the UE on whether it reports DFS / OOR events for each ASA band according to the requirements of regulatory bodies around the world, no new measurements are defined in the specifications to support 3GPP LTE.

[0021] In another embodiment, instead of reinterpreting the existing measurement reports as DFS / OOR events, new measurement reports and associated procedures are defined to specifically indicate (to E-UTRAN) the presence of hidden terminals or primary users. Any UE connected to a cell in the ASA band can be configured to perform and report this new DFS measurement. The eNodeB RRC layer can configure the UE to report the DFS measurement periodically or in response to a trigger, or periodically and in response to a trigger. The eNodeB can configure the measurement events and associated thresholds and offsets to control the DFS measurement reporting of UEs connected in the ASA band. Thus, the exact measurement procedure can be determined by specification. However, the actions taken by the network can be similar to those in the previous embodiment, including UE handover, SCell reconfiguration, and SCell deactivation. Reporting measurements (instead of binary information) can allow the eNodeB RRM function to learn from historical data and apply its own thresholds to improve protection of primary users. The decision to select a different carrier for a given UE ultimately lies with the eNodeB, since the eNodeB can analyze and combine DFS measurements from various UEs connected to it, but if this decision were made at each UE, the network would have to follow whatever the UE indicates to ensure protection of potential primary users.

[0022] <Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA)> FIG. 3 is a diagram illustrating communication between a UE 300 and an eNodeB 320 according to an example embodiment. The UE 300 may be a mobile phone, a computer, or other wireless network device. The UE 300 includes a processor 306 coupled to a memory 304 and a transceiver 310. The processor 306 may include several processors adapted for various operational tasks of the UE, including signal processing and channel measurement and calculations. The memory stores application software 302 that the processor may execute as instructed by a user, and also stores operational instructions for the UE. The processor 306 is also coupled to input / output (I / O) circuitry 308, which may include a microphone, a speaker, a display, and associated software. The transceiver 310 includes a receiver 312 and a transmitter 314 suitable for wireless communication with the eNodeB 320. The transceiver 310 typically communicates with the eNB 320 via various communication channels. For example, the transceiver 310 sends uplink information to the eNodeB 320 via a physical uplink control channel PUCCH and a physical uplink shared channel PUSCH. Correspondingly, the transceiver 310 receives downlink information from the eNodeB 320 via a physical downlink control channel PDCCH and a physical downlink shared channel PDSCH.

[0023] The base station 320 includes a processor 326 coupled to a memory 324, a symbol processing circuit 328, and a transceiver 330 via a bus 336. The processor 326 and the symbol processing circuit 328 may include several processors adapted for various operational tasks, including signal processing and channel measurement and calculations. The memory stores application software 322 that the processor may execute for a particular user, and also stores operational instructions for the eNodeB 320. The transceiver 330 includes a receiver 332 and a transmitter 334 suitable for wireless communication with the UE 300. The transceiver 330 typically communicates with the UE 300 via various communication channels. For example, the transceiver 330 sends downlink information to the UE 300 via a physical downlink control channel PDCCH and a physical downlink shared channel PDSCH. The transceiver 330 also sends special downlink information to the UE 300 via a physical broadcast channel PBCH, a physical hybrid ARQ indicator channel PHICH, a physical control format indicator channel PCFICH, and a physical multicast channel PMCH. Correspondingly, the transceiver 330 receives uplink information from the UE 300 via a physical uplink control channel PUCCH and a physical uplink shared channel PUSCH.

[0024] In accordance with an exemplary embodiment, an E-UTRAN cell such as eNodeB 320 is used in an unlicensed or ASA band where LTE user equipment shares radio resources with other users of equal priority but who strictly adhere to Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA) procedures / protocols. Because 3GPP Long Term Evolution is specifically designed to operate in licensed spectrum, a fundamental problem exists.

[0025] Referring to FIG. 4A, the situation in the downlink direction is similar to the DFS described with reference to FIG. 2. Here, CSMA / CA is implemented as a non-standard proprietary feature according to an exemplary embodiment. The UE is initialized on the LTE band (400). The eNodeB monitors the CSMA / CA band (402). The eNodeB does not transmit any downlink channel even if it senses an ongoing transmission (404). The eNodeB monitors a timeout criterion (408) and continues to monitor the CSMA / CA band (402). If the ongoing transmission ends before the timeout criterion 408, the eNodeB transmits to the UE on the CSMA / CA band (406). Otherwise, once the timeout criterion has passed, RRC signaling causes the UE to release the CSMA / CA band (410) and initiates a handover (412).

[0026] However, the eNodeB may have to transmit some signal regardless of whether an ongoing transmission is detected. The eNodeB transmits discovery reference signal (DRS) bursts with a periodicity on the order of hundreds of milliseconds. The DRS burst may be only one subframe and contains at least the PSS, SSS, and CRS, so that the UE can discover the cell and perform measurements. In a shared cell ID situation, the CSI-RS may also be transmitted during the DRS occasion. The periodic PSS / SSS transmission also allows the UE to obtain coarse time and frequency synchronization with this cell. On the network side, the RRM measurement report based on the DRS allows the eNodeB to determine whether to configure a cell in a certain unlicensed or ASA band for a given UE. In addition to the DRS, the eNodeB needs to periodically transmit some kind of tracking reference signal (TRS) with a periodicity much shorter than that of the DRS, such as 5 ms or 10 ms. The TRS waveform allows the UE to perform automatic gain control (AGC) and fine time and frequency synchronization (tracking). Such a TRS waveform may be based on an existing CRS waveform, which may provide an additional benefit of being useful for channel state information acquisition in case of a CRS-based transmission mode. Furthermore, the eNodeB may periodically transmit a channel state information reference signal (CSI-RS) that enables channel state information acquisition at the UE in a CSI-RS-based transmission mode. The UE may be configured for CSI measurement and for reporting according to the CSI transmission at the eNodeB.

[0027] Referring again to FIG. 3, it may be preferable not to use any downlink channel with CSMA / CA. For example, the Physical Broadcast Channel (PBCH) is not transmitted in cells in unlicensed or ASA bands. Thus, UEs cannot camp on such cells. Similarly, system information is not transmitted. Therefore, such cells can only be configured as SCells, and the PCell is always configured in licensed spectrum. It may also be beneficial not to transmit the Physical Hybrid ARQ Indicator Channel (PHICH) in unlicensed or ASA spectrum. Alternatively, the UL grant transmitted in DCI may act as an implicit ACK / NACK indication by scheduling a retransmission of a previous UL grant. The Physical Control Format Indicator Channel (PCFICH) may or may not be transmitted in unlicensed or ASA spectrum. If an extended PHICH duration is configured, the Control Format Indicator (CFI) is known through the specification. Similarly, the PCFICH is not required for PDSCH transmission in transmission mode 10 (TM10) scheduled by the enhanced physical downlink control channel (EPDCCH). In cross-carrier scheduled PDSCH transmission, the CFI is known through configuration. In comparison, since the PCFICH is transmitted in the same subframe as the PDCCH, the PCFICH can be transmitted whenever the PDCCH is transmitted. Finally, since the Physical Multicast Channel (PMCH) is semi-statically scheduled by the MBMS Coordination Entity (MCE) on reserved resources, it may be beneficial not to transmit the PMCH in unlicensed or ASA spectrum. Otherwise, for unicast downlink transmission, if the CSMA / CA function in the eNodeB indicates that a given subframe may be used for (E)PDCCH or PDSCH transmission, the eNodeB transmits according to LTE Release 12. In one embodiment, the CSMA / CA function in the eNodeB returns a binary indication.If the CSMA / CA functionality for a given cell indicates BUSY on a given carrier, the eNodeB does not transmit (E)PDCCH or PDSCH to any UE. The eNodeB may continue to transmit other signals or channels according to the above recommendations. Alternatively, if the CSMA / CA functionality for a given cell indicates IDLE on a given carrier, the eNodeB may transmit (E)PDCCH and / or PDSCH transmissions, and the eNodeB transmits according to LTE Release 12.

[0028] Referring to FIG. 4B, uplink operation in the CSMA / CA band is similar to downlink operation. The UE is initialized in the LTE band (400). The UE monitors the CSMA / CA band (420). If the UE senses an ongoing transmission (422), it does not transmit any uplink channel. The UE monitors a timeout criterion (426) and continues to monitor the CSMA / CA band (420). If the ongoing transmission ends before the timeout criterion 426, the UE transmits to the eNodeB in the CSMA / CA band (424). Otherwise, once the timeout criterion has passed, the UE sends a BUSY report to the eNodeB (428). RRC signaling causes the UE to release the CSMA / CA band (430) and handover begins (432).

[0029] If the CSMA / CA functionality in the UE indicates that a given subframe cannot be used for uplink transmission, it may be beneficial to discard any configured Sounding Reference Signal (SRS) transmissions so as not to interfere with ongoing transmissions. It may also be beneficial not to transmit a Physical Uplink Control Channel (PUCCH) in unlicensed or ASA spectrum. In this case, PUCCH is transmitted on the PCell only in licensed spectrum. If PUCCH transmissions are allowed in unlicensed or ASA spectrum, In this case, several UE behaviors are expected.

[0030] In some cases, the UE follows existing UE procedures for PUCCH transmission, regardless of the indication of the CSMA / CA function in the UE as to the subframe in which the PUCCH transmission is scheduled. Generally, collisions with ongoing transmissions are unavoidable and the PUCCH may not be properly received at the eNodeB.

[0031] Alternatively, the UE may base any PUCCH transmission on an indication of the CSMA / CA function in the UE for the subframe in which the PUCCH transmission is scheduled: if the CSMA / CA function in the UE indicates BUSY, the UE does not transmit the PUCCH in that subframe, otherwise, if the CSMA / CA function in the UE indicates IDLE, the UE transmits the PUCCH as scheduled.

[0032] The same principles can be applied to the Physical Uplink Shared Channel (PUSCH). In one embodiment, the UE follows existing UE procedures for PUSCH transmission, regardless of the indication of the CSMA / CA function in the UE, for the subframe in which the PUSCH transmission is scheduled. Generally, collisions with ongoing transmissions are unavoidable and the PUSCH may not be properly received at the eNodeB.

[0033] Alternatively, the UE may base any PUSCH transmission on an indication of the CSMA / CA function in the UE for the subframe in which the PUSCH transmission is scheduled. If the CSMA / CA function in the UE indicates BUSY, the UE does not transmit the PUSCH in that subframe. Otherwise, if the CSMA / CA function in the UE indicates IDLE, the UE transmits the PUSCH as scheduled.

[0034] As with DFS, hidden terminals must be taken into account. The above solutions for PUSCH and PUCCH transmissions concern the UE behavior when the CSMA / CA functionality in the UE indicates BUSY for a subframe in which a PUSCH / PUCCH transmission is scheduled. In the case of a hidden terminal with a waveform that is detectable in the UE but not in the eNodeB, the eNodeB may continue to schedule the UE. If the UE follows regular LTE Release 12 operation, this may result in poor performance for the eNodeB-to-UE link and for the links to and from the hidden terminal. This is because the respective transmissions may continue to collide, potentially resulting in excessive interference, so that (a) reliable communication is no longer feasible or (b) at least an acceptable quality of service (QoS) is no longer provided. The opposite case, where the UE does not transmit on PUSCH or PUCCH in a subframe when the CSMA / CA functionality in the UE indicates BUSY, may also result in poor performance, since packets and HARQ ACK / NACK transmissions in BUSY subframes are discarded. Theoretically, the DFS scheme described above can be reused so that the UE informs the eNodeB of a BUSY state of a cell or carrier, so that the MAC (or RRC) layer of the eNodeB can schedule the UE on a different CC to avoid further collisions. Thus, instead of a "DFS event" triggered by the DFS function, the CSMA / CA function may indicate BUSY, but these procedures can be reused if not. However, the time scale for DFS is generally much longer than that for LBT, as in the case of CSMA / CA. Therefore, in the exemplary embodiment, another procedure is provided to deal with hidden terminals in the case of CSMA / CA.

[0035] One objective of the exemplary embodiments is for the upper layers of the UE to inform the upper layers of the eNodeB about the indication of the UE's CSMA / CA capability in a subframe in which the UE is scheduled for uplink transmission. If the UE's CSMA / CA capability indicates IDLE, this state is not notified to the upper layers of the eNodeB, since the UE may always follow the existing LTE Release 12 specifications. Therefore, in some embodiments, actions that the upper layers of the eNodeB (e.g., the eNodeB MAC scheduler) may take in a subframe in which a PUSCH or PUCCH transmission is scheduled and the UE's CSMA / CA capability indicates BUSY are provided.

[0036] It is preferable to use PHY or MAC layer mechanisms since the overall system performance and especially the user-perceived UE throughput are maximized, as the eNodeB can take earlier action by avoiding scheduling the UE on carriers occupied by hidden terminals, where the former has lower latency than the latter. First, to reduce latency, assume that the UE is already configured with up to five serving cells (FIG. 1) on the corresponding component carriers. According to an exemplary embodiment, the serving cells are ordered in ascending order based on ServCellIndex configured by RRC signaling, without excluding other ordering or addressing mechanisms. Then, the four serving cells excluding the PCell are assigned the symbols {00, 01, 10, 11} such that the serving cell (SCell) with the smallest ServCellIndex corresponds to 00, the serving cell (SCell) with the second smallest ServCellIndex corresponds to 01, and so on. If less than four SCells are configured, the unused symbols (e.g., {01, 10, 11} if one SCell is configured) are reserved. Other mappings are not excluded. To ensure the shortest latency, L1 (PHY) signaling is introduced to inform higher layers in the eNodeB about a BUSY indication from the UE's CSMA / CA function in the subframe where a PUSCH or PUCCH transmission is scheduled.

[0037] Therefore, a new PUCCH format is introduced that is always transmitted for the PCell. This new PUCCH format is exactly the same as the existing PUCCH format 1b. However, instead of representing ACK / ACK, ACK / NACK, NACK / ACK, and NACK / NACK / DTX, the QPSK symbols encode the four serving cell indices {00, 01, 10, 11}. For convenience, this new PUCCH format is called format 1c. The eNodeB receiver may distinguish between PUCCH formats 1b and 1c by code division multiplexing, so that these two PUCCH formats may share the same time and frequency resources. Alternatively, this new PUCCH format may have its own time and frequency resources in the PUCCH region. When PUCCH capacity is not an issue, such as in the case of small cells, CDM is preferred for improved spectral efficiency. If the CSMA / CA function in the UE indicates BUSY in a subframe where a PUSCH or PUCCH transmission is scheduled, the UE does not transmit PUSCH or PUCCH as scheduled, but instead indicates a BUSY indication (to the eNodeB) via a PUCCH format 1c transmission on the PCell. Several UE behaviors are possible, but all are assumed to schedule only one SCell at a time to prevent ambiguity at the eNodeB when a PUCCH format 1c is received.

[0038] In one embodiment, PUCCH format 1c indicates which serving cell a BUSY indication occurred. For example, the eNodeB may schedule an uplink transmission in subframe n+k (where k>0) via a UL grant in the DCI received in subframe n. Just before an uplink transmission is scheduled to occur, the CSMA / CA function in the UE starts sensing the medium and indicates to the upper layers of the UE whether the medium is IDLE or BUSY. If IDLE is indicated, the UE proceeds with the scheduled transmission according to the received DCI. If BUSY is indicated, the UE ignores the DCI that schedules the intended uplink transmission and instead sends a PUCCH format 1c for the PCell that encodes (in QPSK symbols) the serving cell where the collision occurred.

[0039] Since the eNodeB was expecting a PUSCH or PUCCH transmission for a particular serving cell, the PUCCH format 1c transmission does not actually convey any additional information to the upper layers of the eNodeB. Therefore, in a different embodiment, the CSMA / CA function in the UE senses all configured serving cells before the scheduled uplink transmission. If IDLE is indicated for the serving cell for which a transmission is scheduled, the UE proceeds with the scheduled transmission according to the received DCI. If BUSY is indicated, the UE ignores the DCI that schedules the intended uplink transmission and instead sends a PUCCH format 1c on the PCell that encodes (in QPSK symbols) the serving cell for which the CSMA / CA function in the UE indicated IDLE. This does not guarantee that the corresponding serving cell is IDLE in the following subframe n+k2 (k2>k), but at least the eNodeB does not continue to schedule uplink transmissions to the same serving cell.

[0040] Introducing the new PUCCH format 1c requires the eNodeB receiver to monitor this new PUCCH format. Therefore, MAC layer procedures may be preferred over the PHY procedures mentioned above. However, sending the MAC control element requires the UE to obtain available uplink resources in addition to the uplink resources that are left unused by not transmitting PUCCH or PUSCH because the medium is BUSY. Also, the time to prepare a PUSCH transmission carrying MAC CE may take longer, so carrier sensing must occur much earlier than for the new PUCCH format, which increases the probability that the CSMA / CA function in the UE indicates IDLE, but the medium is BUSY during subframe n+k. The latency may be even longer if the UE has to send a scheduling request (SR) to transmit an AMC CE. Nevertheless, the MAC layer procedures may still have benefits. For example, this may eliminate the need to impose the constraint that only one SCell is scheduled at a time. Instead, one octet (8 bits) in the MAC CE may be used to simultaneously code all four SCells. The maximum four serving (SCells) are reordered in ascending order based on ServCellIndex, represented by {00, 01, 10, 11}, so that the serving cell (SCell) with the lowest ServCellIndex corresponds to 00, the serving cell (SCell) with the second lowest ServCellIndex corresponds to 01, and so on. Also, the eight bits in one octet of MAC CE correspond to the four SCells with the following mapping: the first two bits correspond to the serving cell represented by {00}, the third and fourth bits correspond to the serving cell represented by {01}, the fifth and sixth bits correspond to the serving cell represented by {10}, and the last two bits correspond to the serving cell represented by {11}, although other mappings and orderings are not excluded.If a bit in a position corresponds to itself, this indicates that the corresponding serving cell was indicated as IDLE. Otherwise, the indication is BUSY and these two bits indicate the serving cell to which the eNodeB should switch. Thus, the bit position in an octet encodes the serving cell to which the bit in that position belongs, and these bits themselves encode the same information transmitted in the above PUCCH format 1c for one cell. For example, the octet {00010011} means that the first, second and fourth serving cells were IDLE and that a transmission for the third serving cell should be transmitted to the first serving cell.

[0041] Modifications are possible in the described embodiments, and other embodiments are possible, within the scope of the claims.

Claims

1. A base station, Initializing a user equipment (UE) in a primary serving cell (PCell) in a licensed frequency spectrum; configuring the UE to communicate with a secondary serving cell (SCell) operating on a carrier of a shared frequency spectrum; monitoring the shared frequency spectrum by a base station to determine whether the shared frequency spectrum is BUSY; transmitting to the UE on the shared frequency spectrum when the shared frequency spectrum is not busy; waiting a first time when the shared frequency spectrum is BUSY; instructing the UE to release the carrier in the shared frequency spectrum after the first time if the shared frequency spectrum is BUSY; receiving a radio resource management (RRM) report from the UE in response to the UE detecting a hidden user on the shared frequency spectrum; 23. A base station comprising: one or more processors configured to:

2. 2. The base station according to claim 1, The base station, wherein the shared frequency spectrum is an unlicensed frequency spectrum.

3. 2. The base station according to claim 1, The base station, wherein the one or more processors are further configured to access the shared frequency spectrum using Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA).

4. 2. The base station according to claim 1, The one or more processors are further configured to instruct the UE to release the shared frequency spectrum after the first time period via radio resource control (RRC) signaling.

5. 2. The base station according to claim 1, The base station, wherein the one or more processors are further configured to instruct the UE to deactivate the SCell after the first time period by medium access control (MAC) signaling.

6. 2. The base station according to claim 1, the one or more processors: transmitting to the UE on a first plurality of channels when the shared frequency spectrum is not busy; excluding a second plurality of channels from the shared frequency spectrum; The base station further configured.

7. 2. The base station according to claim 1, The one or more processors are further configured to exclude a Physical Broadcast Channel (PBCH) from transmission in the shared frequency spectrum.

8. 2. The base station according to claim 1, The one or more processors are further configured to transmit, by the eNB, a discovery reference signal on the shared frequency spectrum when the shared frequency spectrum is BUSY.

9. 2. The base station according to claim 1, The one or more processors are further configured to transmit, by the eNB, a tracking reference signal on the shared frequency spectrum.

10. 2. The base station according to claim 1, a transmitter coupled to the one or more processors; a memory coupled to the one or more processors; The base station further comprises:

11. A user equipment (UE), comprising: Initializing a UE in a primary serving cell (PCell) in a licensed frequency spectrum; Transmitting data from the UE to at least one secondary serving cell (SCell) operating in a shared frequency spectrum; monitoring the at least one SCell by the UE to determine a BUSY state; Transmitting to a base station on the at least one SCell when the at least one SCell is not BUSY; When the at least one SCell is BUSY, reporting the BUSY state of the at least one SCell to the base station; Detecting hidden users on the shared frequency spectrum; transmitting a radio resource management (RRM) report in response to detecting a hidden user on the shared frequency spectrum; 13. A UE comprising: one or more processors configured to:

12. 12. The UE of claim 11, The UE, wherein the shared frequency spectrum is an unlicensed frequency spectrum.

13. 12. The UE of claim 11, The UE, wherein the one or more processors are further configured to access the shared frequency spectrum using Carrier Sense Multiple Access with Collision Avoidance (CSMA / CA).

14. 12. The UE of claim 11, The UE, wherein the one or more processors are further configured to report the BUSY state in an uplink control information packet on a physical uplink control channel of the PCell.

15. 12. The UE of claim 11, The UE, wherein the BUSY state is transmitted on physical uplink control channel resources that are semi-statically configured by radio resource control signaling from the base station.

16. 12. The UE of claim 11, The UE, wherein the BUSY state is transmitted on physical uplink control channel resources that are dynamically signaled by downlink control information packets from the base station.

17. 12. The UE of claim 11, the one or more processors: Determining that the state of the at least one SCell is BUSY; determining an IDLE state of at least one other secondary serving cell; transmitting an identity of the at least one further secondary serving cell to the base station on a physical uplink control channel of the PCell; The UE is further configured.

18. 12. The UE of claim 11, a transmitter coupled to the one or more processors; a memory coupled to the one or more processors; The UE further includes:

19. A base station, Initializing a user equipment (UE) in a primary serving cell in a licensed frequency spectrum; configuring the UE to communicate with a secondary serving cell (SCell) over a shared frequency spectrum; receiving a radio resource management (RRM) report from the UE in response to the UE detecting a hidden user on the shared frequency spectrum; 23. A base station comprising: one or more processors configured to:

20. 20. The base station of claim 19, A base station, wherein the shared frequency spectrum is an authorized shared access (ASA) frequency spectrum and the primary user has a higher access priority than the SCell.

21. 20. The base station of claim 19, A base station, wherein the RRM report is periodic.

22. 20. The base station of claim 19, The base station, wherein the RRM report is a dynamic frequency selection (DFS) event.

23. 20. The base station of claim 19, a transmitter coupled to the one or more processors; a memory coupled to the one or more processors; The base station further comprises:

Images