Outer Loop Link Adaptation for Fast SINR Recovery After Resumption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Wireless communication devices experience throughput penalties due to high error rates during network interruptions or long fades, leading to poorly selected modulation and coding schemes based on historical error estimates rather than actual channel conditions.
Innovation Solution
Implementing a window-averaged block error rate and adjusting the smoothing factor to quickly converge to recent error values, allowing for improved modulation and coding scheme selection by applying offsets to the reported SINR values and reducing the forgetting factor during resumption from interruptions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the base station uses historical block error rate estimates to determine modulation and coding schemes, then the MCS selection is based on accumulated error information, but the MCS allocation remains penalized even after the UE returns from tune-away or fade periods, causing downlink throughput degradation
Solution Approach 1:
The patent applies dynamics by making the BLER estimation adaptive rather than static. The system dynamically adjusts the weighting between historical and recent BLER measurements based on communication continuity. When the UE is actively communicating, historical BLER estimates are used; when the UE tunes away or experiences fades, the system transitions to relying primarily on recent BLER measurements, allowing the MCS allocation to adapt to current channel conditions rather than remaining penalized by historical poor conditions
Solution Approach 2:
The patent changes the parameter weighting in the BLER estimation formula. Instead of using a fixed smoothing factor, the system adjusts the weighting between historical BLER and recent BLER measurements based on whether the UE is currently active or has been absent. This parameter change allows the system to transition from relying on historical averages to relying on recent measurements, thereby improving MCS selection accuracy and downlink throughput after tune-away periods
2Speed
If the base station uses instantaneous block error rate values for MCS mapping, then the MCS selection responds quickly to current channel conditions, but the MCS allocation is overly sensitive to temporary error spikes during tune-away or fade periods
Solution Approach 1:
The patent dynamically changes the smoothing parameter in the BLER estimation based on communication continuity. During active communication, a higher smoothing factor is used to respond quickly to channel changes. During tune-away or fade periods, the system reduces the smoothing factor weight for historical measurements, preventing temporary error conditions from unduly influencing MCS selection while still maintaining some historical context
Solution Approach 2:
The system transitions from a static BLER estimation approach to a dynamic one that adjusts its responsiveness based on UE activity state. When the UE is active, the system is more responsive to current channel conditions. When the UE is absent, the system reduces its sensitivity to temporary errors, creating a dynamic adaptation speed that matches the communication context
Data Source
AI summary
Outer loop link adaptation for device resumption. A user equipment (UE) and base station (BS) may be in communication in a first network (e.g., an LTE network). Communication between the UE and the BS may be interrupted, e.g., due to a long fading environment, the UE tuning away to a second network (e.g., a CDMA network). Accordingly, the measured error rate may increase dramatically. After resumption from the interruption, a negative offset may be applied to a reported SINR value from the UE due to the previous increase in error rate. Upon improvement in the error rate, a larger, positive offset adjustment may be added to the negative offset, allowing the estimated SINR to return to reported SINR more quickly. Additionally, the error rate estimation may be adjusted to converge to a more recently measured more quickly by decreasing a feedback filter coefficient.