MPEG-1 Layer2双声道定点音频编码器的硬件设计  

Two-channel Audio Encoder with Fixed-point of MEPG-1 Layer2 by Hardware

作　　者：史伟伟[1] 赵光东[1] 徐渊[1] 易子林 周德福[1] 

机构地区：[1]深圳大学信息工程学院,深圳518000

出　　处：《计算机与数字工程》2017年第7期1444-1447,1463,共5页Computer & Digital Engineering

摘　　要：MEPG-1 Layer2音频编解码在数字视音频广播领域有着广泛的应用。现有的MEPG-1 Layer2音频编码方案大多在DSP/CPU平台上用软件方式实现,算法本身实现起来复杂,且浮点数的运算本身占用更多的CPU资源,要实现实时性更多的依赖CPU的性能。论文采用全硬件的设计方法,设计实现了MEPG-1 Layer2双声道音频编码器,简化了算法的复杂度且实现了定点数的运算,使用Xilinx提供的ISE完成verilog代码的编写与综合,第三方软件Modelsim完成各功能模块的仿真,在VC707上完成验证,和MEPG-1 Layer2的音频软件编码相比实时性更好,资源使用更少。MEPG-1 Layer2 audio coding and decoding has a wide use in the field of DAB and BVB. The majority of existingMEPG-1 Layer2 audio coding methods are realized by software on DSP/CPU platform. Those approaches are complex to implement,while the floating-point operations consume more CPU resources. Whats worse,to achieve real-time performance mainly dependson the performance of CPU. In our algorithm,hardware implement is adopted and two-channel audio encoder of MEPG-1 Layer2 ispresented and realized. It simplifies the complexity of the traditional algorithms and implements the fixed-point number operations.The verilog coding and synthesis is completed by ISE provided by Xilinx and the simulation of each function module is done via mod-elsim. The algorithm is verified in VC707.The experimental result demonstrates that the presented method holds better real-time per-formance and fewer resources are employed compared to the previous algorithms.

关 键 词：MEPG-1Layer2 MODELSIM 实时 VERILOG 

分 类 号：TN912.3[电子电信—通信与信息系统]