Joint Detection Accelerator for TD-SCDMA Signal Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current joint detection systems in TD-SCDMA wireless communication face challenges in efficiently processing signals from multiple users, leading to interference and reduced performance, especially at low signal-to-noise ratios, due to the complexity of channel estimation and interference cancellation.

Innovation Solution

A joint detection accelerator is designed to perform joint detection of received signals using a programmable DSP and hardware circuitry, including a complex multiply accumulate unit, simplified complex multiply accumulate unit, and normalized floating point divider, which reduces memory accesses and power consumption while improving processing speed and precision through pre-scaling and pipelined architecture.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If joint detection processing is performed using traditional software-based methods on a host processor, then flexibility and programmability are maintained, but processing speed and real-time performance deteriorate due to the computational complexity of channel estimation and interference cancellation

Engineering Contradiction:
Improveprocessing speedVSAvoidcomputational complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The patent replaces the software-based processing mechanism with a hardware-based joint detection accelerator. The accelerator uses dedicated hardware circuits including complex multiply-accumulate units, simplified complex multiply-accumulate units, and normalized floating-point dividers to perform joint detection operations in hardware rather than software, thereby achieving real-time processing of multiple user signals while maintaining the ability to handle computational complexity through parallel hardware operations

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The joint detection accelerator is segmented into multiple specialized processing units: complex multiply-accumulate units for general computations, simplified complex multiply-accumulate units for specific operations where one complex value is restricted to ±1 or ±j, and normalized floating-point dividers. This segmentation allows each unit to be optimized for its specific function, improving overall processing speed while managing computational complexity through division of labor

Inventive Principle:
Principle #1Segmentation

2Measurement precision

If memory accesses are increased to store intermediate results during joint detection, then processing accuracy is improved, but power consumption increases

Engineering Contradiction:
Improvedata extraction accuracyVSAvoidpower consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent implements different data path configurations with varying precision characteristics. The complex multiply-accumulate units and simplified complex multiply-accumulate units provide different levels of computational precision for different stages of processing. By using higher precision where needed and optimized precision where sufficient, the system maintains data extraction accuracy while reducing unnecessary power consumption from uniform high-precision operations throughout

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The accelerator dynamically adjusts operational parameters including the use of simplified complex multiply-accumulate units for operations where one value is restricted to ±1 or ±j, and normalized floating-point dividers for division operations. These parameter changes optimize the balance between precision and power consumption by matching computational resources to the actual requirements of each processing stage

Inventive Principle:
Principle #35Parameter changes

3Productivity

If chip operational modes are extended to reduce sleep duration, then processing throughput is improved, but power consumption increases

Engineering Contradiction:
Improveprocessing throughputVSAvoidpower consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The joint detection accelerator is designed with dynamic power management capabilities that allow the chip to switch between active processing modes and low-power sleep modes. The hardware architecture supports rapid state transitions, enabling the system to maintain high processing throughput during active periods while efficiently entering sleep modes between processing tasks, thereby improving overall productivity without sustaining high power consumption continuously

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP2067268B1Architecture for joint detection hardware accelerator
Publication Date: 2017.07.26 MEDIATEK INC
  • EP2067268B1 patent drawingFigure 1
  • EP2067268B1 patent drawingFigure 2
  • EP2067268B1 patent drawingFigure 3

AI summary

A joint detection system is configured to perform joint detection of received signals and includes a joint detection accelerator and a host processor. The joint detection accelerator may include a memory unit to store input data values, intermediate results and output data values; one or more computation units to process the input data values and the intermediate results, and to provide output data values to the memory unit; a controller to control the memory and the one or more computation units to perform joint detection processing; and an external interface to receive the input data values from the host processor and to provide output data values to the host processor. The computation units may include a complex multiply accumulate unit, a simplified complex multiply accumulate unit and a normalized floating point divider. The memory unit may include an input memory, a matrix memory, a main memory and an output memory.