Distributed high-precision multi-scale signal analyzer data synchronous acquisition method

Abstract

The invention discloses a distributed high-precision multi-scale signal analyzer data synchronous acquisition method, and belongs to the technical field of testing. According to the method, the requirement for network resource consumption is low, the algorithm stability is higher, the signal analyzer can be supported to quickly achieve high-precision time synchronization in a higher-efficiency and lower-resource-consumption mode, then the purpose of data acquisition is achieved, and a distributed collaborative testing system is better supported; according to the method, the processes of triggering and sending the synchronous message for coupling are separated on the time axis, and the synchronous message with self phase information is sent at random time in a period, so that the communication load is reduced, the channel congestion is avoided, and the stability of the method is improved.

Description

Technical Field

[0001] The invention belongs to the field of testing technology, and in particular relates to a distributed high-precision multi-scale signal analyzer data synchronization acquisition method. Background Art

[0002] As an electronic testing technology, distributed testing is widely used in distributed equipment performance testing, regional electromagnetic spectrum monitoring, communications and other scenarios. Existing distributed testing systems are divided into distributed testing systems based on wired networks and wireless networks, each with its own advantages and disadvantages. The wired network distributed testing system supports high-speed data transmission and efficient processing, and supports parallel control of multiple distributed test terminals. However, due to the limitations of wired network cables, the deployment range is limited and can only be used within a certain area. The wireless network distributed testing system solves the problem of limited deployment range while retaining the advantages of the limited network distributed testing system. However, whether it is a wired network or a wireless network distributed system, the efficiency and accuracy of data synchronization acquisition are affected because the network channel is easily interfered with.

[0003] As the foundation of distributed test systems, data synchronization acquisition technology is the key to determining the measurement accuracy of distributed test systems. In existing distributed test systems, the implementation methods of data synchronization acquisition technology can be divided into two types: hardware methods and software methods. Most hardware methods use GPS synchronization modules to control data acquisition. The advantage is that the synchronization accuracy is high, but it is necessary to integrate GPS modules in the test terminal, which increases the hardware cost. In addition, GPS technology is subject to the control of the United States and is not easy to obtain. The domestic Beidou Satellite Navigation System (BDS) has not yet been commercialized on a large scale. Software methods mainly use the Network Time Protocol (NTP) and the Precision Time Protocol (PTP) to achieve synchronous data acquisition. The accuracy of the NTP protocol is in the hundreds of milliseconds, and the PTP protocol can reach the sub-nanosecond level. However, in actual application, both the NTP and PTP protocols require the network to be a peer-to-peer network. In the case of a complex network, the synchronization accuracy cannot be guaranteed.

[0004] In the existing distributed test system, the implementation methods of data synchronization acquisition technology can be divided into two types: hardware method and software method. Most hardware methods use GPS synchronization module to control data acquisition, which has the advantage of high synchronization accuracy, but it needs to integrate GPS module in the test terminal, which increases the hardware cost. In addition, GPS technology is controlled by the United States and is not easy to obtain. The domestic Beidou satellite navigation system (BDS) has not yet been commercialized on a large scale. Software methods mainly use network time protocol (NTP) and precision time protocol (PTP) to achieve synchronous data acquisition. The accuracy of NTP protocol is in the hundreds of milliseconds, and PTP protocol can reach sub-nanosecond level. However, both NTP and PTP protocols require the network to be a peer-to-peer network, which is almost impossible in practical applications, so it is difficult to guarantee synchronization accuracy. Summary of the invention

[0005] In view of the above technical problems existing in the prior art, the present invention proposes a distributed high-precision multi-scale signal analyzer data synchronization acquisition method, which has a reasonable design, overcomes the shortcomings of the prior art, and has good effects.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A distributed high-precision multi-scale signal analyzer data synchronization acquisition method comprises the following steps:

[0008] Step 1: Phase calculation;

[0009] Step 2: Synchronous message sending;

[0010] Step 3: Delay compensation;

[0011] Step 4: Frequency offset compensation;

[0012] Step 5: Synchronize message processing and cache update;

[0013] Step 6: Multi-scale RFA (Reachback Firefly Algorithm) phase adjustment;

[0014] After the longest timing cycle of the multi-phase space ends, the node settles the phase difference of each layer of the buffer area and calculates the starting phase matrix of the next cycle, that is, the phase adjustment matrix with the buffer as input, to avoid the impact of data packet disorder caused by communication delay, effectively reduce the amount of hardware calculation, and finally realize the synchronization of distributed signal analyzer nodes, and repeat steps 1 to 6.

[0015] Preferably, in step 1, a multi-scale phase space is used. The multi-scale phase space belongs to a discrete space, and its phase calculation or phase growth method is step-like. The calculation formula is as follows:

[0016]

[0017]

[0018] Where i represents the i-th node in the distributed network, Ph i represents the normalized phase value of the ith node, l represents the phase space series, k il represents the quantization level of node i in the lth level phase space, Δt l represents the update time interval of the l-th level phase space, T l represents the time cycle of the l-th level phase space, t n+l represents the n+l moment, k l(t n ) indicates t n The quantization level of the phase space at the first level at the moment k l (t n+1 ) indicates t n+l The quantization level of the l-th level phase space at time;

[0019] That is, a self-growth method is adopted, and every Δt l The discrete phase value of the node at level l is increased by 1 at a time, and the phase vector only needs to be updated once every minimum time resolution.

[0020] Preferably, in step 2, the process of triggering and sending synchronization messages for coupling is separated on the time axis, and the two tasks are independent of each other; during the operation, the node sends a synchronization message with its own phase information at a random time within a cycle; the time of sending the synchronization message is random rather than when the node phase reaches a threshold, and the synchronization message contains the phase information of the node at the moment of sending rather than simple pulse information; the random message sending mechanism belongs to the event coupling mode, which does not require continuous exchange of phase information between nodes, reducing the communication load; when the node receives the synchronization message, it extracts the phase information from it, filters out the valid phase information by comparing it with its own phase, and processes the synchronization message according to the steps in the synchronization message processing and cache update.

[0021] Preferably, in step 3, the network communication delay is divided into uncertain delay and deterministic delay; for the uncertain delay, the MAC layer SFD timestamp is used to eliminate it, that is, after the SFD (Start Frame Delimiter) interrupt of the node sending the trigger information packet is generated, the local time count at this time is quickly added to the end of its sending queue, and accordingly, the receiving node quickly records its local time count in the interrupt service after receiving the SFD of the trigger packet, and finally parses the received trigger packet; the deterministic delay includes transmission delay, propagation delay, and reception verification delay; the deterministic delay is obtained by querying the technical data of the RF chip used or by calibration experiments; dynamic compensation is performed based on the obtained uncertain delay and deterministic delay.

[0022] Preferably, in step 4, frequency offset compensation is mainly caused by a low-performance RC oscillator inside a communication chip in a wireless network; in order to reduce errors and improve synchronization accuracy, the true frequency of the RC oscillator of the node used is first calibrated, then compared with the theoretical frequency, and finally the adjusted true minimum phase scale is calculated based on the theoretical minimum phase scale.

[0023] Preferably, in step 5, the node receives the synchronization message sent by the neighbor node within the single-hop communication range during the synchronization process, and the coupled synchronization processing is the core of the entire synchronization mechanism; the node processes the phase data of the synchronization data packet using a hierarchical similar phase weighting method, that is, in each layer, only the remote phase information that does not meet the refractory period condition and is closest to the local phase is retained.

[0024] Preferably, in step 6, the cache area used by the acquisition method only caches one phase difference information, that is, the minimum phase difference information received in the same cycle. Merging nodes with smaller phase differences can speed up the synchronization process and make the synchronization process immune to interference from some abnormal nodes.

[0025] Beneficial technical effects brought by the present invention:

[0026] The present invention proposes a data synchronization acquisition method suitable for a signal analyzer as a distributed collaborative test node. Compared with the existing method, the method has less requirements on network resource consumption and higher algorithm stability, and can support the signal analyzer to quickly achieve high-precision time synchronization in a more efficient and less resource consuming manner, thereby achieving the purpose of data acquisition, so as to better support the distributed collaborative test system.

[0027] The present invention separates the process of triggering and sending synchronization messages for coupling on the time axis, and sends out synchronization messages with their own phase information at random times within a cycle, thereby reducing communication load, avoiding channel congestion, and improving the stability of the method. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 It is a block diagram of the distributed collaborative test data synchronization acquisition solution;

[0029] Figure 2 This is a block diagram of the distributed high-precision multi-scale synchronization algorithm solution. DETAILED DESCRIPTION

[0030] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0031] The present invention mainly aims at the data synchronization acquisition problem of the distributed collaborative test system, and designs a distributed high-precision multi-scale signal analyzer data synchronization acquisition method. The method comprehensively considers the clock synchronization accuracy, network resource requirements, algorithm stability, etc., to achieve the purpose of high-precision time synchronization in a more efficient and less resource consuming manner, so that the signal analyzer can synchronize the signal data more stably and accurately, so as to better support the distributed collaborative test system. The data synchronization acquisition scheme of the present invention is as follows: Figure 1 As shown, the synchronization algorithm is as follows Figure 2 As shown, it is explained in detail below.

[0032] Step 1: Phase calculation. The distributed high-precision multi-scale data synchronous acquisition method designed by the present invention adopts a multi-scale phase space, which belongs to a discrete space. Its phase calculation or phase growth method is step-like, and the calculation formula is as follows:

[0033]

[0034]

[0035] Where i represents the i-th node in the distributed network, Ph i represents the normalized phase value of the ith node, l represents the phase space series, k il represents the quantization level of node i in the lth level phase space, Δt l represents the update time interval of the l-th level phase space, T l represents the time cycle of the l-th level phase space, t n+l represents the n+l moment, k l (t n ) indicates t n The quantization level of the phase space at the first level at the moment k l (t n+1 ) indicates t n+l The quantization level of the l-th level phase space at time;

[0036] That is, a self-growth method is adopted, and every Δt l The discrete phase value of the node at level l is increased by 1 at a time, and the phase vector only needs to be updated once every minimum time resolution.

[0037] Step 2: Sending synchronization message. The process of triggering and sending synchronization messages is separated on the timeline, and the two tasks are independent of each other. During operation, the node sends a synchronization message with its own phase information at a random time within a cycle. The time of sending the synchronization message is random rather than when the node phase reaches the threshold, and the synchronization message must contain the phase information of the node at the moment of sending rather than simple pulse information. This mechanism belongs to the event coupling mode, which does not require continuous exchange of phase information between nodes, reducing the communication load. When the node receives the synchronization message, it extracts the phase information from it, filters out the valid phase information by comparing it with its own phase, and processes the synchronization message according to the steps in synchronization message processing and cache update.

[0038] Step 3: Delay compensation. Network communication delay is divided into uncertain delay and deterministic delay. For uncertain delay, the present invention adopts MAC layer SFD timestamp technology to eliminate it, that is, after the SFD interrupt of the node sending the trigger information packet is generated (the conflict detection has passed at this time), the local time count at this time is quickly added to the end of its sending queue. Correspondingly, the receiving node quickly records its local time count in the interrupt service after receiving the SFD of the trigger packet, and finally parses the received trigger packet. Because both timestamps are stamped at the moment when an interrupt service is added after the SFD transmission is completed, there is no influence of the transmission time. In theory, the delay of the two timestamps only differs by the propagation time and the time difference between the two interrupt service functions. In the IEEE802.11 protocol, the chip often does not provide an SFD terminal or SFD output pin. At this time, the sending or receiving interrupt can be used instead, and then the true value of the timestamp is calculated according to the interrupt timing in the chip manual. Deterministic delay is mainly composed of transmission delay, propagation delay, and reception verification delay. Deterministic delay can be obtained by querying the technical data of the RF chip used or by calibration experiments. Therefore, dynamic compensation is performed based on the obtained uncertain delay and deterministic delay.

[0039] Step 4: Frequency offset compensation. Frequency offset compensation is mainly caused by the low-performance RC oscillator inside the communication chip in the wireless network. In order to reduce errors and improve synchronization accuracy, the real frequency of the RC oscillator of the node used is first calibrated, then compared with the theoretical frequency, and finally the adjusted real minimum phase scale is calculated based on the theoretical minimum phase scale.

[0040] Step 5: Synchronization message processing and cache update. During the synchronization process, the node receives the synchronization message sent by the neighbor node within the single-hop communication range, and the coupled synchronization processing is the core of the entire synchronization mechanism. The present invention stipulates that the node uses a hierarchical close phase weighting method to process the phase data of the synchronization data packet, that is, only the remote phase information that does not meet the refractory period condition and is closest to the local phase is retained in each layer.

[0041] Step 6: Multi-scale RFA phase adjustment. Due to the processing capacity limitation of network communication nodes, there is a lag in the adjustment after analysis, and the wireless data packets are easily received out of order. To address this problem, the buffer area in the method proposed in the present invention only caches one phase difference information, that is, the minimum phase difference information received in the same cycle. Firstly, merging nodes with smaller phase differences can speed up the synchronization process, make the synchronization process immune to the interference of some abnormal nodes, save hardware resources, and reduce the coupling calculation intensity.

[0042] After the longest timing cycle of the multi-phase space ends, the node settles the phase difference of each layer of the buffer area and calculates the starting phase matrix of the next cycle, that is, the phase adjustment matrix with the buffer as input, to avoid the impact of data packet disorder caused by communication delay, effectively reduce the amount of hardware calculation, and finally realize the synchronization of distributed signal analyzer nodes, and repeat steps 1 to 6.

[0043] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A distributed high-precision multi-scale signal analyzer data synchronization acquisition method, characterized in that: The steps include: Step 1: Phase calculation; Step 2: Synchronous message sending; Step 3: Delay compensation; Step 4: Frequency offset compensation; Step 5: Synchronize message processing and cache update; Step 6: Multi-scale RFA phase adjustment; After the longest timing cycle of the multi-phase space ends, the node settles the phase difference of each layer of the buffer area, calculates the starting phase matrix of the next cycle, that is, the phase adjustment matrix with the buffer as input, realizes the synchronization of the distributed signal analyzer nodes, and repeats steps 1 to 6.

2. The distributed high-precision multi-scale signal analyzer data synchronization acquisition method according to claim 1 is characterized in that: In step 1, a multi-scale phase space is used. The multi-scale phase space belongs to a discrete space, and its phase calculation or phase growth method is step-like. The calculation formula is as follows: Where i represents the i-th node in the distributed network, Ph i represents the normalized phase value of the ith node, l represents the phase space series, k il represents the quantization level of node i in the lth level phase space, Δt l represents the update time interval of the l-th level phase space, T l represents the time cycle of the l-th level phase space, t n+l represents the n+l moment, k l (t n ) indicates t n The quantization level of the phase space at the first level at the moment k l (t n+1 ) indicates t n+l The quantization level of the l-th level phase space at time; Using the self-increment method, every Δt l The discrete phase value of the node at level l is increased by 1 at a time, and the phase vector only needs to be updated once every minimum time resolution.

3. The distributed high-precision multi-scale signal analyzer data synchronization acquisition method according to claim 1 is characterized in that: In step 2, the process of triggering and sending synchronization messages for coupling is separated on the time axis, and the two tasks are independent of each other; during the operation, the node sends a synchronization message with its own phase information at a random time within a cycle; the time of sending the synchronization message is random rather than when the node phase reaches the threshold, and the synchronization message contains the phase information of the node at the moment of sending rather than simple pulse information; the random message sending mechanism belongs to the event coupling mode, which does not require continuous exchange of phase information between nodes, reducing the communication load; when the node receives the synchronization message, it extracts the phase information from it, filters out the valid phase information by comparing it with its own phase, and processes the synchronization message according to the steps in synchronization message processing and cache update.

4. The distributed high-precision multi-scale signal analyzer data synchronization acquisition method according to claim 1, characterized in that: In step 3, the network communication delay is divided into uncertain delay and deterministic delay. For the uncertain delay, the MAC layer SFD timestamp is used to eliminate it, that is, after the SFD interrupt of the node sending the trigger information packet is generated, the local time count at this time is quickly added to the end of its sending queue. Correspondingly, the receiving node quickly records its local time count in the interrupt service after receiving the SFD of the trigger packet, and finally parses the received trigger packet. Deterministic delay includes transmission delay, propagation delay, and reception verification delay. The deterministic delay is obtained by querying the technical data of the RF chip used or through a calibration experiment; dynamic compensation is performed based on the obtained uncertain delay and deterministic delay.

5. The distributed high-precision multi-scale signal analyzer data synchronization acquisition method according to claim 1, characterized in that: In step 4, frequency offset compensation is mainly caused by the low-performance RC oscillator inside the communication chip in the wireless network. In order to reduce errors and improve synchronization accuracy, the real frequency of the RC oscillator of the node used is first calibrated, then compared with the theoretical frequency, and finally the adjusted real minimum phase scale is calculated based on the theoretical minimum phase scale.

6. The distributed high-precision multi-scale signal analyzer data synchronization acquisition method according to claim 1, characterized in that: In step 5, during the synchronization process, the node receives the synchronization message sent by the neighboring node within the single-hop communication range, and the coupled synchronization processing is the core of the entire synchronization mechanism; the node processes the phase data of the synchronization data packet using a hierarchical similar phase weighting method, that is, in each layer, only the remote phase information that does not meet the refractory period condition and is closest to the local phase is retained.

7. The distributed high-precision multi-scale signal analyzer data synchronization acquisition method according to claim 1, characterized in that: In step 6, the buffer area used in the acquisition method only caches one phase difference information, that is, the minimum phase difference information received in the same cycle. Merging nodes with smaller phase differences can speed up the synchronization process and make the synchronization process immune to interference from some abnormal nodes.

Images