A Reconfigurable Neural Network ASIC for Detector Front-End Data Compression at the HL-LHC

• Published in: IEEE transactions on nuclear science Vol. 68; no. 8; pp. 2179 - 2186
• Main Authors: Guglielmo, Giuseppe Di, Fahim, Farah, Herwig, Christian, Valentin, Manuel Blanco, Duarte, Javier, Gingu, Cristian, Harris, Philip, Hirschauer, James, Kwok, Martin, Loncar, Vladimir, Luo, Yingyi, Miranda, Llovizna, Ngadiuba, Jennifer, Noonan, Daniel, Ogrenci-Memik, Seda, Pierini, Maurizio, Summers, Sioni, Tran, Nhan
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.08.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Acidity; Algorithms; Application specific integrated circuits; application-specific integrate circuit (ASIC); Application-specific integrated circuit (ASIC); artificial intelligence (AI); Artificial neural networks; autoencoder; CMOS; Compression; Computer architecture; Data compression; Data transmission; Design optimization; Detectors; Energy consumption; Field programmable gate arrays; Gates (circuits); hardware accelerator; High level synthesis; high-level synthesis (HLS); INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; Integrated circuits; Ionizing radiation; Large Hadron Collider; Large Hadron Collider (LHC); Learning algorithms; Machine learning; machine learning (ML); Neural networks; Particle physics; PHYSICS OF ELEMENTARY PARTICLES AND FIELDS; Power consumption; Radiation; Sensors; single-event effect (SEE) mitigation; Sodium channels; Solenoids; Task analysis; Training
• ISSN: 0018-9499; 1558-1578
• DOI: 10.1109/TNS.2021.3087100

Despite advances in the programmable logic capabilities of modern trigger systems, a significant bottleneck remains in the amount of data to be transported from the detector to off-detector logic where trigger decisions are made. We demonstrate that a neural network (NN) autoencoder model can be implemented in a radiation-tolerant application-specific integrated circuit (ASIC) to perform lossy data compression alleviating the data transmission problem while preserving critical information of the detector energy profile. For our application, we consider the high-granularity calorimeter from the Compact Muon Solenoid (CMS) experiment at the CERN Large Hadron Collider. The advantage of the machine learning approach is in the flexibility and configurability of the algorithm. By changing the NN weights, a unique data compression algorithm can be deployed for each sensor in different detector regions and changing detector or collider conditions. To meet area, performance, and power constraints, we perform quantization-aware training to create an optimized NN hardware implementation. The design is achieved through the use of high-level synthesis tools and the hls4ml framework and was processed through synthesis and physical layout flows based on a low-power (LP)-CMOS 65-nm technology node. The flow anticipates 200 Mrad of ionizing radiation to select gates and reports a total area of 3.6 mm 2 and consumes 95 mW of power. The simulated energy consumption per inference is 2.4 nJ. This is the first radiation-tolerant on-detector ASIC implementation of an NN that has been designed for particle physics applications.