Low-dose CT scans decrease patient radiation dose but add considerable noise and loss of high-frequency information. Current deep learning methods deal with denoising or super resolution individually, while joint restoration is less explored. We present a combination of wavelet transform and pix2pix model, a hybrid system that utilizes discrete wavelet decomposition to provide multi-resolution, frequency-aware inputs to a supervised conditional GAN. Each low-dose slice is resolved through Daubechies-1 wavelet decomposition into LL, LH, HL, and HH subbands, concatenated into a four-channel tensor, and paired with its high-quality counterpart. pix2pix generator is trained using a PatchGAN adversarial loss and an L reconstruction loss to realize simultaneous noise reduction and resolution enhancement. We evaluated our experiment by comparing SSIM and PSNR against standalone pix2pix models and found considerable improvement, which is supported by qualitative output as well by showing improved anatomical structure preservation.