DSP-based mixed-signal solution solves advanced audio processing challenges

Since the idea of ​​the 1970s, especially in the past decade, data signal processing (DSP) has become a familiar technology driven by the proliferation of media applications. Beginning with military voice and image processing, this technology first developed into an audio and video application environment, and then entered the more mainstream electronics field.

This article refers to the address: http://


Today, designers are facing a series of signal processing challenges: the need for applications with advanced functionality, portability and low energy consumption is increasing. The audio field is a typical application: from professional headphones to electronic stethoscopes and remote sensors, new applications are emerging, requiring engineers to use battery power to achieve audio compression and recording, sound management, and back at relatively low cost and miniaturized design. Wave elimination and speech processing.


But the portability of the application adds to the difficulty of these challenges: low energy consumption not only requires increased battery life, but also requires a reduction in battery size and weight. Size limitations also require fewer components as possible, and the complexity of the device continues to increase as the market demands increased product features. Finally, the rapid changes in the market put pressure on designers, who must complete projects more quickly than before and turn their ideas into sales as quickly as possible. Therefore, IP multiplexing or some form of programmability is critical.


There is ample evidence that the driving force for signal processing integration has also impacted the separation between traditional analog and digital electronics, making IC design for mixed signals more common. For chip designers, this is inherently difficult: for example, they need to design components with multiple ground and power supplies, and ensure adequate isolation between the analog and digital parts, as some of them can interfere with the other.


Despite these inherent challenges, the needs of system designers have made it difficult to integrate analog and digital technologies on a chip. The industry has solved this problem in many different ways: manufacturers of linear circuits have begun to add digital capabilities to their devices; and vendors of programmable logic technology have also tried to integrate analog functions. However, perhaps the most important and successful thing is that manufacturers of microprocessors, DSPs, and DSP cores have struggled to integrate more peripheral functions on the chip. Today, equipment vendors with true expertise and processing technology can provide single-chip signal processing solutions with ADCs, DACs, display drivers, power management, integrated amplifiers, and many peripheral support features commonly used in special applications.


Single chip audio processing solution


Figure 1 shows a schematic of an optimized mixed-signal system-on-chip audio processing solution developed by AMI Semiconductor. The chip, called the BelaSignaTM 250, provides an excellent example of the level of integration that is currently possible with the latest mixed-signal semiconductor processing technology.


Based on a parallel processing architecture that combines exceptional power efficiency with high performance, the device forms a complete audio solution that includes ADCs and DACs, filters, amplifiers and power management circuitry. Using a fully programmable 16-bit fixed-point dual Harvard architecture DSP core as its main processor means that the BelaSigna 250 can be quickly configured in software—using the library features provided by AMIS or third-party software partners to make this The process is accelerated.


The parallelism introduced by incorporating the Weighted Overlay (WOLA) filter bank coprocessor drives digital filtering by performing conversions on one signal between the time domain and the frequency domain and other vector-based calculations. This hardware acceleration feature makes power efficiency and space requirements in portable designs possible.


This parallel processor architecture is surrounded by a full stereo audio signal chain, including inputs via a 16-bit oversampling sigma-delta A/D converter and differential digital power outputs including analog line outputs and direct drive amplifiers. The outstanding fidelity of the audio chain provides an overall system dynamic range of 88dB. The ability of the architecture to perform double precision operations means that this signal fidelity can be maintained during processing. Very low system noise and low group delay are achieved by the new time-frequency conversion algorithm used in the WOLA filter bank coprocessor.


The signal data within the device is managed by an audio optimized DMA unit, an input/output processor (IOP). This unit manages the collection of data from the ADC and input FIFO. After processing, it also feeds digital data to the audio output stage and output FIFO. Each FIFO itself has two internal interfaces: one is a standard order, when the data arrives, the address of the most recent input block changes; the second is a "smart" interface for time domain filtering, the most recent input is It remains the same there.


In addition to the flexibility of the interface connection, the data in the FIFO can be accessed in four general different ways: mono mode; simple stereo mode; digital mixing mode; and full stereo mode. These modes offer a variety of options for stereo data and cross or block storage for mono/stereo input/output configurations.


By combining two main processing units, hardware acceleration, data management, and input/output signal conditioning in a single device, designers can integrate larger systems faster and easier. They can significantly reduce the number of components to meet the small size requirements inherent in portable equipment. Adding I2S, PCM, SPI, I2C, TWSS and GPIO direct interfaces, as well as the device's small form factor (64-pin LFBGA package, size is only 7mmx7mm) and integrated IOP for processing data, further help designers to achieve miniaturization.


In addition, the BelaSigna 250 adds many additional features to help reduce power consumption. The device has a low power design. For example, the system operates at a 1.8V supply voltage and a 20MHz clock frequency under average load and typically consumes less than 5mA. For additional power savings, each input channel can be turned off via software, while a low current standby mode that requires only 0.05mA of current supply extends battery life.


Electronic Stethoscope Reference Design


One of the most promising areas of integrated mixed-signal device applications is in the medical electronics arena: handheld devices and portable patient monitors, among others, are particularly versatile for portability, low power consumption and rapid development. Particularly striking is the electronic stethoscope. The device is beginning to be recognized and offers significant potential advantages in the assessment and classification of cardiopulmonary diseases compared to previous traditional products.


To help manufacturers take advantage of this opportunity, AMI Semiconductor has developed an electronic stethoscope reference design (ESRD) based on the BelaSigna 250 (Figure 2). The design is designed to enable rapid development and deployment of products with industry-leading audio performance and battery life. It offers three frequency response modes, while an adjustable amplifier up to 21dB allows the practitioner to set the volume with 10 levels of control. In addition, the device can record heart sounds and lung sounds in non-volatile memory that can be played back at normal speed or half speed to aid in diagnosis.


This reference design allows selection of buttons between “ring” (low frequency), “vibration” (high frequency) and extended frequency mode. Developers can easily implement additional features because the BelaSigna 250 is a programmable platform: for example, they can introduce heart rate monitoring or introduce adaptive processing solutions that separate the fetal heartbeat from the maternal heartbeat. In addition, the device includes an interface to connect directly to wireless and wired communication chips: this allows us to develop connectivity for data transfer to a PC or other device via the Bluetooth® protocol in the future, for later use. Data is stored, analyzed and restored.

Portable Projector Bluetooth

Shenzhen Happybate Trading Co.,LTD , https://www.szhappybateprojectors.com