Theses and Dissertations
Permanent URI for this collectionhttp://ir.daiict.ac.in/handle/123456789/1
Browse
2 results
Search Results
Item Open Access Neural Network Architectures for Integrated Circuits(Dhirubhai Ambani Institute of Information and Communication Technology, 2023) Nagrani, Khyati; Maiti, Tapas KumarThis thesis presents the architecture design and implementation of neural networks(NNs) for integrated circuit design. The architecture consists of adders,multipliers, and rectified linear unit (ReLU) blocks. Three architectures, namely,Single-In Single-Out (SISO), Multiple-In Single-Out (MISO), and Multiple-In Multiple-Out (MIMO) are developed. In neural networks, weight values are necessaryand they are supplied from a memory source. The weight values were preparedby training the NNs model on software. Finally, the SISO, MISO, and MIMOneural-networks were taped out. These architectures can be used for intelligentco-processor development.Item Open Access Active sub-harmonic mixer(Dhirubhai Ambani Institute of Information and Communication Technology, 2015) Majithiya, Bhumik Kiritbhai; Gupta, SanjeevA superheterodyne receiver is commonly used receiver, but its main problem is its large size. To increase the integration and reduce the size of the receiver, the concept of direct conversion receiver came in the picture. It converts the RF signal into baseband signal (baseband is the signal which has very low frequency) directly in first mixing only. But the major disadvantage is LO (Local Oscillator) self mixing which create DC offset. To solve this issue, the sub-harmonic mixer was introduced. The sub-harmonic mixer presented here is designed using Gilbert cell mixer. Two Gilbert cell mixers are cascaded to facilitate mixing RF signal with second harmonic of the LO signal.